Focused analgesia in waking and hypnosis: Effects on pain,
memory, and somatosensory event-related potentials
Vilfredo De Pascalis *, Immacolata Cacace, Francesca Massicolle
University of Rome ``La sapienza'', Department of Psychology, Via dei Marsi 78, 00185 Rome, Italy
Received 2 March 2007; received in revised form 29 August 2007; accepted 7 September 2007
Abstract
Somatosensory event-related potentials (SERPs) to painful electric standard stimuli under an odd-ball paradigm were analyzed in
12 high hypnotizable (HH), 12 medium hypnotizable (MH), and 12 low hypnotizable (LH) subjects during waking, hypnosis, and a
cued eyes-open posthypnotic condition. In each of these conditions subjects were suggested to produce an obstructive imagery of
stimulus perception as a treatment for pain reduction. A No-Analgesia treatment served as a control in waking and hypnosis conditions.
The subjects were required to count the number of delivered target stimuli. HH subjects experienced significant pain and
distress reductions during posthypnotic analgesia as compared to hypnotic analgesia and between these two analgesic conditions
as compared to the two control conditions. Outside of hypnosis, these subjects remembered less pain and distress levels than they
reported during hypnotic and posthypnotic analgesia treatments. In contrast, for waking-analgesia treatment, HH subjects remembered
similar pain and distress levels to those they reported concurrently with the stimulation. HH subjects, during hypnotic and
posthypnotic analgesia treatments, detected a smaller number of target stimuli and displayed a significant amplitude reduction
of the midline frontal and central N140 and P200 SERP components. No significant SERP differences were observed for these subjects
between treatments in waking condition and between hypnotic and posthypnotic analgesic treatments. For the MH and LH
subjects no significant N140 and P200 amplitude changes were observed among analgesic conditions as compared to control conditions.
These amplitude findings are seen as indicating that hypnotic analgesia can affect earlier and later stages of stimulus
processing.
Ó 2007 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.
Keywords: Pain; Hypnosis; Analgesia; Pain sensation; Pain memory; Somatosensory event-related potentials; N140; P200
1. Introduction
Previous findings have shown that obstructive hallucinations
of noxious stimulation in hypnosis reduce pain
sensation (for reviews see, [8,30]) and the amplitude of a
later P300 component of the SERPs [14,15,48], indicating
that the locus of hypnotic influence is not in the
initial sensory experience itself, but rather in the cognitive-emotional
component of the information processing.
However, more recent EEG findings have also
evidenced that focused analgesia, in hypnosis, may
reduce a stimulus-locked 40 Hz-EEG synchronization
response [12] that is believed to reflect perceptual aspects
of stimulation [1]. Therefore, the main aim of the present
study was to further evaluate the modulatory effect of
hypnotic analgesia on both the earlier N140 and later
P300 component of the SERPs, the former believed to
be more stimulus orientated and the latter expression
of the ongoing cognitive-emotional processing (e.g.,
[26]). The study was devoted to address a very important
and controversial question concerning the impact of
hypnosis on the response to suggestions (see [38]). This
was carried out by comparing the effects of an analgesia
suggestion, administered during a non-hypnotic waking
0304-3959/$32.00 Ó 2007 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.pain.2007.09.005
*
Corresponding author. Tel.: +39 0649917643; fax: +39 064451667.
E-mail address: v.depascalis@caspur.it (V. De Pascalis).
www.elsevier.com/locate/pain
Pain 134 (2008) 197­208
condition, with those obtained delivering the same suggestion
just after the induction of hypnosis and, later,
after a cued posthypnotic condition that is believed to
produce a deeper hypnosis [2].
A typical characteristic of the deeply hypnotized
individual is the appearance, spontaneously upon
emerging from hypnosis, of an apparent amnesia for
that which had occurred while in the hypnotic state
(posthypnotic amnesia). There seems to be a general
agreement that posthypnotic amnesia is not the same
as simple forgetting [27,44], and that it is a product
of hypnotic suggestion, direct or implicit [32­
34,37,52], and also a product of dissociation [21,60].
No reports are known to us evaluating the effect of
spontaneous posthypnotic distortion of pain memory.
Thus, a further purpose of the present study was to
evaluate, upon emergence from hypnosis, the degree
in which retrospective pain and distress ratings are
spontaneously distorted, and whether these distortions
are modulated by individual differences in hypnotizability.
The rationale for this evaluation is based on
results of previous studies showing that memory distortion
of pretreatment pain contributes to an exaggeration
of self reports of pain relief [13,22,29,39,40].
Further aim of the present investigation was to evaluate
whether: (1) in HH individuals, the analgesic suggestions
are more effective following a formal hypnotic
induction (hypnotic and posthypnotic conditions) than
in the waking condition; (2) pain reduction during hypnotic
and posthypnotic analgesia conditions is accompanied
by an attenuation of the N140 and P200 peaks of
the SERPs.
2. Methods
2.1. Subjects
Thirty-six right-handed undergraduate students (18 women
and 18 men; age range 19­28 yr) were selected for high
(N = 12; 6 women and 6 men), medium (N = 12; 6 women
and 6 men), and low (N = 12; 6 women and 6 men) levels of
hypnotic susceptibility. The subjects were tested using the Italian
translation by [57] of the Stanford Hypnotic Susceptibility
Scale, Form C (SHSS:C; [16,59]). They were categorized as
being HH subjects (N = 12, M = 10.5, SD = 0.7) or LH ones
(N = 12, M = 2.9, SD = 1.5) when their scores on SHSS:C
were, respectively, 1 SD above or below the group mean of a
larger group of subjects (N = 105, M = 6.8, SD = 3.9; 65
women and 40 men). The moderately hypnotizable group
was formed with subjects who showed hypnotizability scores
1 SD within the group mean (N = 12, M = 6.3, SD = 0.8).
Two female hypnotists carried out the assessment of hypnotic
susceptibility about 20 days prior to the EEG recording session.
In this session, hypnosis was induced again by one of
the two hypnotists who did not know the hypnotizability level
of the subject. Subjects were admitted to participate in the
experiment only if they reported an absence of medication
use or medical conditions that might interfere with pain sensitivity
(e.g., diabetes mellitus, high blood pressure, heart diseases,
asthma, post trauma to hands, frostbite, arthritis,
Raynauďs syndrome). Subjects were not informed of their
hypnotic ability and of the relevance of hypnotic ability in
the experiment. Women who were in a menstrual period were
invited for EEG recordings in another occasion.
2.2. Procedure
The selected subjects were individually invited in the EEG
lab and upon arrival they were informed about the nature of
the painful electric stimulation. In accordance with the ethical
norms of the Italian Association of Psychology (AIP), a written
consent was obtained if they agreed to participate in the
experiment. In this session, hypnosis was induced for the second
time using an Italian translation of the original American
protocol of the Stanford Hypnotic Clinical Scale (SHSC; [42]).
The subjects were not informed about their hypnotizability
level during the EEG recording session and were all nai¨ve
volunteers.
2.3. Pain treatments
During waking and hypnosis conditions, the following five
pain treatments were administered to each subject: (1) WakingPain
(No-Treatment: W-Pain). Subject was required to detect
target painful stimuli (eyes-open) without giving suggestions
to reduce pain. (2) Waking-Analgesia (W-Analgesia). Suggestion
to produce an obstructive imagery of stimulus perception
by imagining a glove that was covering the finger and the hand
and focusing on sensation in the finger and hand and experiencing
that all sensations of the stimulated finger will be
attenuated (eyes-open). (3) Hypnosis-Pain (No-Treatment:
Hy-Pain). At the end of hypnotic induction and hypnotic
testing, painful stimuli were delivered without suggestions to
reduce pain (eyes-closed). (4) Hypnosis-Analgesia (Hy-Analgesia).
In hypnosis condition was given an analgesia suggestion
as in (2) (eyes-closed). (5) Post Hypnosis-Analgesia (P.HyAnalgesia).
Just before getting out from hypnosis, the subject
was suggested that he/she will tend to sink deeper and deeper
into involvement in the hypnotic state with open eyes, after
that the experimenter will have knocked two times on the wall
of the sound proof box (`fractionation technique', see
Barabasz and Watkins, 2005; p. 193). When the subject was
just out from hypnosis, the experimenter knocked two times
and suggested to the subject that he/she was going into a deeper
hypnosis state and the above-reported analgesia suggestion
was administered. At the end of P.Hy-Analgesia treatment, the
subject was waked up from hypnosis. Both waking and hypnosis
conditions were counterbalanced across subjects in order to
avoid possible order effects or habituation. Within each
waking or hypnosis condition, the order of the treatments
was not varied to prevent for a proactive effect of suggestion.
Between waking and hypnosis conditions, a resting period of
12 min was given. In each condition, the subject was asked
to count the number of delivered target stimuli.
Each treatment condition lasted about 5 min. Painful stimuli
were applied to the subjects middle finger of the right hand
and, at the end of each condition, they were asked to rate any
198 V. De Pascalis et al. / Pain 134 (2008) 197­208
pain and distress experienced for standard stimuli on two separate
10 point numeric rating scales (NRS; [31]). On the left
and right sides of the NRS-sensory scale, there were, respectively,
the descriptors `0 = no pain sensation' and `10 = the
most intense pain sensation imaginable'. Similarly, for the
NRS-distress scale, the descriptors ranged from `0 = not at
all distressfuľ to `10 = the most distress imaginable'. An involuntariness
measure of pain reduction effect was obtained at the
end of Post-Hypnosis-Analgesia condition by requiring participants
to rate on the NRS how much pain reduction they experienced
as occurring involuntarily (from `0 = quite voluntarily'
to `10 = absolutely involuntarily').
Approximately 5 min after the waking or hypnosis session
ended, the participants used two 10-point NRS scales to rate
remembered pain and distress sensations they experienced for
each experimental condition.
2.4. Sensory and pain thresholds
Somatosensory stimuli were delivered by applying two silver­silver
chloride cup electrodes to the palmar surfaces of
the distal and medial phalanges of the middle finger of the
right hand. The electrode cup (1 cm in diameter) was filled with
an electro-conductive hypoallergic cream and impedance was
kept below 30 kX. Stimuli were unipolar electrical pulses
(2-ms duration) generated by a constant current stimulator
(Digimiter, Mod DS7A).
Sensory and pain thresholds were determined for each participant
just before the EEG recording. Participants received
first a series of single, unipolar pulses separated by 10 s intervals,
starting with an intensity of .05 mA and increasing with
steps of .05 mA until the participant reported that the minimum
detectable stimulus was reached. The intensity level of
the just noticeable pin-prick was taken as the sensory threshold
associated with ascending intensity levels. A similar, but
reversed procedure, was used to obtain the sensory threshold
associated with descending intensity levels, starting from
.5 mA and decreasing with steps of .05 mA. The two thresholds
were then averaged to obtain the participanťs sensory
threshold. After the sensory threshold, pain tolerability threshold
was determined by delivering stimuli of increasing intensity
(steps of 0.5 mA) until a very painful pin-prick was reported.
After this level, stimulus intensity was then increased until
the subject reported the delivered stimulus as unbearable. This
value was defined as the greatest level of pain which a subject is
prepared to tolerate. Stimulus intensity used during experimental
conditions was 0.5 mA under this tolerability level.
2.5. Electric stimulation
In each experimental condition subjects were engaged in an
oddball task consisting of 70 electrical stimuli wherein infrequent
targets (14.5%) were interspersed among frequently
occurring standard stimuli (85.5%).
Target stimuli were presented in a pseudo-randomized
order so that at least 2 standard stimuli preceded a target
one. The inter-stimulus interval was set at a constant time of
3 s. Each standard stimulus consisted of one unipolar pulse
with a duration of 2 ms. Target stimulus was formed by pairing
two standard stimuli with an inter-pulse interval of 25 ms.
2.6. EEG acquisition and processing
The EEG was recorded by using pure tin electrodes
mounted on an Electro-cap [5] placed on frontal (F3, F4), parietal
(P3, P4) and midline (Fz, Cz, Pz) scalp sites. Linked earlobes
served as reference with a forehead ground. Electrode
impedance was kept below 3 kX and raw EEG signals were
recorded using nine amplifiers (amplifier gains set at 10,000
with a band pass of 0.5­75 Hz). Eye movement (EOG) was
recorded in a bipolar arrangement, superior orbit referenced
to the outer canthus of the left eye. Trials on which the EEG
or the EOG exceeded 100 lV were rejected automatically.
The EEG was acquired in digital form, using an IBM-compatible
computer, by sampling at 1024 Hz per channel with a
12-bit resolution (Metrabyte Dash-16). For each instruction
condition, 70 epochs (60 for standard and 10 for target stimuli)
were digitized and stored on hard disk, using a time period of
1000 ms. For each recording epoch, a 100 ms period before
stimulus onset was taken as baseline. EEG sweeps with artifacts
due to scalp muscle or stimulus contamination, head or
electrode movement, slow potential variations, false positive
button pressing on standard non-target stimuli were a posteriori
eliminated. In this study, only EEG sweeps corresponding
to standard stimuli were off-line analyzed. This was done since
there is experimental evidence that standard stimuli are more
probable to elicit a `pain-specific' SERP response (a potential
which is usually considered of the same family, but with a
shorter latency, of the classic P300 wave) that is more dependent
from nociceptive component of the stimulus and less
influenced by the ongoing cognitive information processing
not intrinsic to pain [3,4]. Each EEG epoch was low-pass filtered
at 15 Hz (FIR filter 3 dB, 12 dB/octave roll-off) and then
averaged.
The most stable and reliable SERP peaks were a negative
component peaking at about 140 ms (N140: 142.8  11.2 ms)
and a positive component peaking at about 200 ms (P200:
198  10.6 ms). A peak amplitude measure of the N140 component
was obtained as the minimum peak amplitude detected
within a 70­200 ms time interval. An amplitude measure of the
P200 component was obtained as the maximum peak amplitude
within a 120­250 ms time window. These peak values
were obtained by using algorithms of the ASYST-Keithley
programming system for the calculus of local maxima and
minima. Time windows used for the measure of N140 and
P200 peaks were chosen after visual inspection of the SERPs.
2.7. Statistical analyses
A simple split-plot ANOVA with Hypnotizability, serving
as between group variables, was used to assess individual differences
on subjective threshold measures.
A repeated measure split-plot ANOVA, comprising Hypnotizability
(3) and Treatment (5), respectively, as between
and within factors, was performed to assess the effect of hypnotizability
and treatments on the number of omissions of target
stimuli.
To test the effect of experimental treatments on retrospective
pain and distress ratings, a repeated measure split-plot
ANOVA was carried out, comprising Hypnotizability (3), Recollection
(2) and Treatment (5), respectively, as between and
V. De Pascalis et al. / Pain 134 (2008) 197­208 199
within factors. The Recollection factor included both contingent
pain ratings and remembered pain ratings as a within factor
with nested Treatment as a repeated measure factor. A
similar ANOVA was used to assess the distress and remembered
distress ratings.
To test the impact of hypnotic depth on the response to
analgesia suggestion change scores of both pain and distress
were calculated for each waking and hypnosis condition by
subtracting concurrent ratings during the suggestion of analgesia
from those obtained in the correspondent control condition.
Change scores were analyzed using separate ANOVAs
with Hypnotizability (3) serving as between-subjects factor
and Condition (3; Waking, Hypnosis, Post-hypnosis) as
within-subjects factor.
For each N140 and P200 peak amplitude measure, two separate
repeated measures ANOVAs were carried out. One
ANOVA was performed for quadrant recording sites (F3,
F4, and P3, P4) to assess hemispheric and anterior/posterior
effects [46] using Hypnotizability (3) as between-subjects factors,
and Treatment (5), Recording Site (2; frontal, parietal),
and Hemisphere (2) as within-subjects factors. A separate
ANOVA was carried out to assess N140 and P200 changes
for midline recording sites (Fz, Cz, and Pz).
Huynh-Feldt epsilon correction of significance levels was
used when necessary [58]. Post hoc comparisons were carried
out by using a t-test procedure with Bonferroni correction of
a = .05 [36]. SAS-8.02 was used for all statistical analyses.
3. Results
3.1. Threshold measures, involuntariness ratings, omitted
targets, and remembered pain intensity and distress
ratings
The ANOVA on sensory threshold, pain, and distress
thresholds failed to yield any significant effect (all Fs
were not significant).
The ANOVA on involuntariness ratings yielded main
effects for Hypnotizability, F(2, 33) = 6.35, MSe = 6.92,
P = .005, indicating that the analgesic effect of the suggestion
was experienced as occurring more automatically
in HH subjects as compared to medium and LH
ones (7.6, 4.7, and 3.8, respectively).
The ANOVA carried out for the number of omitted
targets across treatments yielded a main effect for Hypnotizability,
F(2, 33) = 5.12, MSe = 4.47, P = .012, indicating
that HH subjects had a greater mean number of
omissions as compared to MH and LH subjects (1.75,
.93, and.53, respectively). Moreover, the main effect
for Treatment was highly significant, F(3.7,
123.3) = 9.58, MSe = 2.09, e = .934, P < .0001, as it
was the case for the Treatment and Hypnotizability
interaction, F(7.5, 123.3) = 6.75, MSe = 2.09, e = .934,
P < .0001. Post-hoc comparisons of the means indicated
that for HH subjects the number of omitted targets was
more pronounced during Hy-Analgesia and P.Hy-Analgesia
as compared to W-Pain (t = 3.7, P < .01, and
t = 6.5, P < .0001). The t test also indicated that there
were more omissions during Hy-Analgesia and P.HyAnalgesia
in comparison with Hy-Pain treatment
(t = 4.9, P < .001, t = 5.6 P = 0.001, respectively). The
increase in the number of omitted targets was also significant
for the P.Hy-Analgesia as compared to the
Hy-Analgesia treatment (t = 4.3, P < .001). The mean
number of omitted targets during treatments for the
hypnotizability groups is reported in Table 1.
The ANOVA performed on both contingent and
recalled pain ratings with Recollection and Treatment
as within factors and Hypnotizability as a between factor
yielded a significant main effect for Treatment,
F(2.9, 95.0) = 10.19, MSe = 2.22, e = .722, P < .0001,
and a significant interaction between Hypnotizability
and Treatment, F(5.8, 95.0) = 5.04, MSe = 2.22,
e = .722, P = .0002. Post-hoc comparisons of the overall
pain means displayed that in HH participants there were
significant reductions of both experienced and remembered
pain during Hy-Analgesia and P.Hy-Analgesia
as compared to W-Pain and to W-Analgesia treatment
(W-Pain vs Hy-Analgesia and P.Hy-Analgesia: t = 4.6
and t = 4.95, P < .001; W-Analgesia vs Hy-Analgesia
and P.Hy-Analgesia: t = 2.4, P < .05 and t = 3.9,
P < .01). The t-tests showed that there was a significant
reduction in experienced and remembered pain scores
from Hy-Pain to Hy-Analgesia and P.Hy-Analgesia
treatments (t = 4.6, and t = 5.1, P < .001), and a significant
reduction in pain measures during P.Hy-Analgesia
as compared to Hy-Analgesia treatment (t = 3.2,
P < .01). For HH participants, differences in pain ratings
between W-Pain and W-Analgesia and between
the latter and Hy-Pain treatment were not significant
(t = .9, and t = 1.4, P > .05, respectively), while, for
these subjects, the reduction of pain during Hy-Pain,
with respect to W-Pain treatment, was near the significance
level (t = 2.1, P = .06; see Fig. 1).
Table 1
Mean number and standard deviation (SD) of omitted target stimuli
for high, medium, and low hypnotizable participants (HH, MH, and
LH) during waking (W) and hypnosis (Hy) treatments
Hypnotizability N Treatment Mean SD
HH 12 W-Pain 0.33 0.65
W-Analgesia 0.83 1.40
Hy-Pain 0.42 1.16
Hy-Analgesia 2.17 1.27
P.Hy-Analgesia 5.00 2.59
MH 12 W-Pain 0.60 1.72
W-Analgesia 0.64 1.62
Hy-Pain 0.67 1.15
Hy-Analgesia 1.25 1.96
P.Hy-Analgesia 1.50 1.93
LH 12 W-Pain 0.50 1.00
W-Analgesia 0.50 1.45
Hy-Pain 0.75 2.60
Hy-Analgesia 0.58 0.99
P.Hy-Analgesia 0.33 0.89
200 V. De Pascalis et al. / Pain 134 (2008) 197­208
Pain reductions obtained for LH participants did not
reach the significance level (all ts < 2, P > .05). The MH
participants reported smaller pain ratings during
Hy-Analgesia and P.Hy-Analgesia as compared to WPain
(t = 2.3, and t = 2.4, P < .05) and to W-Analgesia
treatment (t = 2.2, and t = 2.4, P < .05).
Furthermore, a significant three-way interaction was
obtained for the Hypnotizability, Recollection, and
Treatment factors, F(6.2, 101.6) = 2.86, MSe = .523,
e = .77, P < .05. This interaction indicated that HH subjects,
for hypnotic analgesia and posthypnotic analgesia
treatments, recollected significantly lower pain sensations
as compared to the pain reported in the concurrent
ratings (t = 2.8, P = .020, and t = 3.1 P = .011, respectively,
for experienced vs concurrent pain ratings of
the Hy-Analgesia, and P.Hy-Analgesia conditions).
The patterns of experienced and remembered pain ratings
across hypnotizability groups are displayed in
Fig. 1.
A similar analysis performed on experienced and
remembered distress ratings showed that the main effect
of Hypnotizability was near to reach the significance
level, F(2, 33) = 2.97, MSe = 35.66, P = .065, indicating
that HH participants tended to have lower distress levels
than MH and LH ones (4.4, 5.5, 6.2, respectively). The
Recollection factor was highly significant, F(1,
33) = 20.73, MSe = 1.32, P < .0001, indicating that all
the subjects, in the Recollection phase, rated a lower
level of distress than they did contingently with treatments.
The Treatment effect was also highly significant,
F(2.8, 92.9) = 15.19, MSe = 2.53, e = .704, P < .0001, as
well as the interaction between Hypnotizability and
Treatment, F(5.6, 92.9) = 7.18, MSe = 2.53, e = .704,
P < .0001. Post-hoc comparisons of the overall pain
means displayed that in HH participants there were significant
reductions of both experienced and remembered
distress scores during Hy-Analgesia and P.Hy-Analgesia
as compared to W-Pain and to W-Analgesia treatment
(W-Pain vs Hy-Analgesia and P.Hy-Analgesia: t = 7.3
and t = 6.9, P < .0001; W-Analgesia vs Hy-Analgesia
and P.Hy-Analgesia: t = 4.2, P < .01 and t = 6.1,
P < .0001). These subjects also displayed a significant
reduction in experienced and remembered distress scores
during Hy-Analgesia and P.Hy-Analgesia treatments as
compared to Hy-Pain (t = 5.9, and t = 6.1, P < .0001),
and a significant reduction in distress measures during
P.Hy-Analgesia as compared to Hy-Analgesia treatment
(t = 3.5, P < .01). In HH subjects distress rating measures
did not differ between W-Pain and W-Analgesia
(t = 1.8, P > .05) and between W-Analgesia and HyPain
treatment (t = .36, P > .05), while these measures
were significantly reduced during Hy-Pain as compared
to W-Pain treatment (t = 2.9, P < .05; see Fig. 2). Distress
reductions, obtained for LH participants, did not
reach the significance level (all ts<2, P > .05). The MH
participants reported smaller distress ratings during
Hy-Analgesia and P.Hy-Analgesia as compared to WPain
(t = 2.4, and t = 2.2, P < .05) and to W-Analgesia
treatment (t = 2.3, and t = 2.4, P < .05).
Finally, a significant three-way interaction was
obtained for the Hypnotizability, Recollection and
Treatment factors, F(6.9, 114.7) = 2.68, MSe = .653,
e = .869, P < .05. This interaction indicated that HH
subjects, for hypnotic analgesia and posthypnotic analgesia
treatments, recollected significantly lower distress
sensations as compared to concurrent painful stimulation
(t = 4.3, P < .010, and t = 4.5, P < .010, respectively,
for experienced vs concurrent pain ratings of
the Hy-Analgesia, and P.Hy-Analgesia conditions).
The trend of this interaction, observed for the HH subjects,
can be derived from the patterns of the experienced
and remembered pain ratings displayed in Fig. 2.
Fig. 1. Means of Experienced Sensory Pain and Remembered Pain with standard errors to electrical stimulation by hypnosis group and condition
(Waking-Pain, Waking-Analgesia; Hypnosis-Pain; Hypnosis-Analgesia; Post-Hypnosis Analgesia). The asterisk (*) indicates significant differences
between experienced and remembered pain scores during analgesic treatments in hypnosis (*P < .05, t test).
V. De Pascalis et al. / Pain 134 (2008) 197­208 201
To test the impact of hypnosis on the response to suggestions,
separate ANOVAs on pain and distress change
scores were performed. A main effect for Hypnotizability
was found for pain change scores, F(2, 33) = 4.90,
MSe = 3.06, P = .014, and for distress change scores,
F(2,33) = 4.98, MSe = 3.72, P = .013. These effects indicated
that HH subjects had more pronounced pain and
distress reductions as compared to MH and LH subjects,
while there were no significant differences between
MH and LH subjects (Pain: 1.39, .43, and .29; Distress:
1.64, .51, and .37, respectively, for HH, MH, and LH
subjects). The Condition main effect was also significant
for pain change scores, F(1.9, 64.8) = 5.59, MSe = 1.34,
e = .982, P = .007, and distress change scores, F(1.9,
62.7) = 5.93, MSe = .97, e = .951, P = .005. Post-hoc
comparison of the means showed that the effect of pain
reduction was significantly stronger during P.Hy-Analgesia
as compared with W-Analgesia and Hy-Analgesia,
while there were no significant differences between the
last two conditions (Pain: 1.15, .34, and .62, respectively;
t = 2.47, P = .019, for P.Hy-Analgesia vs W-Analgesia;
t = 2.17, P = .037, for P.Hy-Analgesia vs Hy-Analgesia;
t = 1.04, P > .05, for Hy-Analgesia vs W-Analgesia).
Similarly, distress reduction was significantly stronger
during P.Hy-Analgesia as compared with W-Analgesia
and Hy-Analgesia, while there were no significant differences
between the last two conditions (Distress: 1.27,
.54, and .72, respectively; t = 2.52, P = .016, for P.HyAnalgesia
vs W-Analgesia; t = 2.73, P = .009, for
P.Hy-Analgesia vs Hy-Analgesia; t = .83, P > .05, for
Hy-Analgesia vs W-Analgesia).
The interaction between Hypnotizability and Condition
was also significant for pain change scores, F(3.9,
64.8) = 2.55, MSe = 1.34, e = .982, P = .048, and distress
change scores, F(3.8, 62.7) = 2.67, MSe = .97,
e = .951, P = .042. These interactions indicated that in
HH subjects the effect of suggestion was more pronounced
and significant during P.Hy-Analgesia as compared
to W-Analgesia (Pain: 2.60 vs .77, t = 3.62,
P = .004; Distress: 2.70 vs .92, t = 4.58, P = 0.0008),
and to Hy-Analgesia (Pain: 2.60 vs 1.04, t = 3.27,
P = .007; Distress: 2.70 vs 1.42, t = 3.15, P = 0.009),
while the difference between Hy-Analgesia and W-Analgesia
was not significant (Pain: 1.04 vs .77, t = .46,
P > .05; Distress: 1.42 vs .92, t = 1.73, P > 0.05). Difference
scores for MH and LH subjects were all less than
0.75 and all comparisons between conditions were
non-significant (all t < 1.5, P > .05).
3.2. Effects of experimental factors on N140 and P200
peak amplitudes
With the onset of the standard electric stimulus we
observed a complex consisting of two components, a
N140 peak that displayed a maximum amplitude at
fronto-central sites (peak latency of 142 ms at Fz and
139 ms at Cz), and a P200 peak that showed a maximum
amplitude at central site (peak latency of
196 ms at Cz).
3.2.1. Quadrant N140 peak amplitude
For quadrant N140 amplitude scores, the ANOVA
yielded the following significant effects: (1) Treatment,
F(1.8, 61.4) = 6.19, MSe = 10.20, e = .465, P = .004;
(2) Recording Site, F(1, 33) = 79.21, MSe = 64.40,
P < .0001; (3) Treatments and Recording Site interaction,
F(2.7, 90.4) = 4.36, MSe = 4.40, e = .684
P = .0118; (4) Hypnotizability and Treatment interaction,
F(3.7, 61.4) = 3.54, MSe = 10.2, e = .465,
P = .013. The first effect indicated that there was a more
pronounced N140 peak during W-Pain and HypnosisPain
as compared to the other three analgesic conditions
Fig. 2. Means of Experienced Distress and Remembered Distress with standard errors to electrical stimulation by hypnosis group and condition
(Waking-Pain, Waking-Analgesia; Hypnosis-Pain; Hypnosis-Analgesia; Post-Hypnosis Analgesia). The two asterisks (**) indicate significant
differences between experienced and remembered pain scores during analgesic treatments in hypnosis (**P < .01, t test).
202 V. De Pascalis et al. / Pain 134 (2008) 197­208
(10.2, 9.5, 10.5, 9.3, and 9.2 lV, respectively, for W-Pain
W-Analgesia, Hy-Pain Hy-Analgesia, and P.Hy-Analgesia;
t values were all significant, P < .05). The second
effect showed a more pronounced peak over frontal sites
as compared to the parietal ones (12.3 vs 7.2 lV). The
third effect disclosed that, for Hy-Analgesia and P.HyAnalgesia,
as compared to Hy-Pain treatment, there
was a smaller N140 peak over frontal and parietal scalp
sites; frontal site: t = 4.2, P < .001 and t = 3.8, P < .001;
parietal site: t = 2.2, P < .05 and t = 3.0, P < .01. In
contrast, there were no significant differences between
W-Analgesia and W-Pain treatments (P > .05). The
fourth effect showed that N140 peak reduction was
mainly due to the HH participants that produced significant
smaller N140 peak amplitudes during Hy-Analgesia
and P.Hy-Analgesia as compared to Hy-Pain (8.0
and 7.7 vs 10.9 lV, respectively; t values were 7.7, and
6.7, P < .0001) and as compared to W-Pain treatment
(8.0 and 7.7 vs 10.2 lV, respectively; t values were 3.1,
and 3.3, P < .01, respectively). In contrast, for these subjects,
the reduction in N140 amplitude during W-Analgesia
as compared to W-Pain did not reach the
significance level (9.0 vs 10.2 lV, respectively, t = 2.0,
P = 0.068). Similarly, for these subjects, there were no
significant differences in N140 amplitude between HyAnalgesia
and P.Hy-Analgesia treatments (t = 1.1
P > .05).
3.2.2. Midline N140 peak amplitude
The ANOVA on midline N140 amplitudes yielded
the same significant effects as those evidenced for quadrant
data: Treatment, F(1.8, 59.0) = 5.07, MSe = 11.15,
e = .447, P = .017; Recording Site, F(2, 66) = 101.04,
MSe = 48.29, e=1.101, P < .0001; Treatment and
Recording Site interaction, F(4.0, 130.9) = 4.37,
MSe = 3.28, e = .496, P = .0025; Hypnotizability and
Treatment interaction, F(3.6, 59.0) = 5.38,
MSe = 11.15, e = .447, P = .0015. Moreover, this analysis
evidenced a significant triple interaction between
Hypnotizability, Treatments, and Recording Site,
F(7.9, 131.5) = 2.53, MSe = 3.23, e = .498, P = .013.
This effect indicated that, across Fz and Cz recording
sites, HH participants had more pronounced N140
amplitude reductions during Hy-Analgesia and P.HyAnalgesia
as compared to Hy-Pain (Fz: 11.4 and 11.7
vs 16.7 lV, respectively; t values were 6.6, and 6.0,
P < .0001; Cz: 12.2 and 12.6 vs 17.4 lV, respectively; t
values were 7.4, and 6.9, P < .0001) and as compared
to W-Pain treatment (Fz: 11.4 and 11.7 vs 15.6 lV,
respectively; t values were 3.5, and 3.2, P < .01, respectively;
Cz: 12.2 and 12.6 vs 15.9 lV, respectively; t values
were 2.8, and 2.5, P < .05, respectively). For HH subjects,
there were no significant differences on midline
N140 amplitude between W-Analgesia and W-Pain (all
t < 2, P > .05) and between Hy-Analgesia and P.HyAnalgesia
treatments (all ts < 1, P > .05). SERP
responses recorded at midline locations are displayed
in Fig. 3.
3.2.3. Quadrant P200 peak amplitude
Quadrant P200 amplitude scores were found to be
modulated by the same significant effects as those evidenced
for the N140 peak: (1) Treatment, F(2.5,
82.2) = 6.02, MSe = 9.19, e = .623, P = .0018; (2)
Recording Site, F(1, 33) = 19.89, MSe = 36.32,
P < .0001; (3) Treatments and Recording Site interaction,
F(3.7, 123.7) = 3.76, MSe = 3.38, e = .937
P = .008; (4) Hypnotizability and Treatment interaction,
F(5.0, 82.2) = 2.63, MSe = 9.19, e = .623,
P = .052. The Treatment effect indicated that this positive
peak was smaller during Hy-Analgesia and P.HyAnalgesia
as compared to Hy-Pain treatment (6.1, 6.5,
and 7.6 lV, respectively; all ts > 3.5, P < .001), while
the difference in peak amplitude between W-Pain and
W-Analgesia did not reach the significance (t = 1.1,
P > .05). The Recording Site effect showed a higher peak
over frontal region as compared to the parietal one (7.9
vs 5.9 lV). The Treatment and Recording Site interaction
effect disclosed that over the frontal region, during
Hy-Analgesia and P.Hy-Analgesia, there were smaller
peaks as compared to Hy-Pain condition, while these
differences were not significant over parietal one (frontal
region: t = 5.0, P < .001 and t = 3.11, P < .01; parietal
region: t = 1.7, P > .051 and t = 1.4, P > .05). The Hypnotizability
and Treatment interaction showed that HH
participants (but not MH and LH participants) produced
significant smaller P200 peak amplitudes during
Hy-Analgesia and P.Hy-Analgesia treatments as compared
to Hy-Pain (5.6 and 6.1 vs 8.1 lV, respectively; t
values were 9.7, P < .0001, and 3.0, P < .01) and as compared
to W-Pain treatment (5.6 and 6.1 vs 8.1 lV; t values
were 3.2, P < .01, and 2.3, P < .05, respectively). For
these subjects, there were no significant differences on
P200 amplitude between W-Analgesia and W-Pain (8.0
vs 8.1 lV; t < 1, P > .05) and between Hy-Analgesia
and P.Hy-Analgesia treatments (5.6 vs 6.1; t < 1,
P > .05).
3.2.4. Midline P200 peak amplitude
The ANOVA on P200 amplitude over midline
recordings yielded the same significant effects as those
evidenced for quadrant data: Treatment, F(3.0,
99.1) = 7.58, MSe = 8.87, e = .751, P < .0001; Recording
Site, F(2, 66) = 50.71, MSe = 51.51, e=1.099,
P < .0001; Treatment and Recording Site interaction,
F(4.5, 147.6) = 2.29, MSe = 2.44, e = .559, P < .0001;
Hypnotizability and Treatment interaction, F(6.0,
99.1) = 3.92, MSe = 8.87, e = .751, P = .0015. Moreover,
a significant triple interaction between Hypnotizability,
Treatments, and Recording Site was also
obtained, F(8.9, 147.6) = 2.59, MSe = 2.44, e = .559,
P = .0085. This effect indicated that, across Fz and Cz
V. De Pascalis et al. / Pain 134 (2008) 197­208 203
recording sites, HH participants had more pronounced
P200 amplitude reductions during Hy-Analgesia and
P.Hy-Analgesia as compared to Hy-Pain (Fz: 8.0 and
8.4 vs 12.6 lV, respectively; t values were 4.5, and 4.6,
P < .001; Cz: 9.6 and 10.2 vs 14.4 lV, respectively; t values
were 4.5, and 4.2, P < .0001) and as compared to WPain
treatment (Fz: 8.0 and 8.4 vs 12.8 lV, respectively;
t values were 4.2, and 3.2, P < .01, respectively; Cz: 9.6
and 10.1 vs 15.4 lV, respectively; t values were 4.5,
and 4.5, P < .001, respectively). For HH subjects, there
were no significant differences, across midline locations,
for P200 amplitude between W-Analgesia and W-Pain
(all ts < 2, P > .05) and between Hy-Analgesia and
P.Hy-Analgesia treatments (all ts < 1, P > .05). SERP
responses recorded at midline locations are displayed
in Fig. 3.
3.3. Relation between pain and distress ratings and ERP
responses
Correlation coefficients, obtained between rating
measures and N140 and P200 peak amplitudes, did
not yield any significant relationship, except for a significant
correlation found between P200 amplitude over
left-parietal scalp location (P3) and pain ratings
obtained during P.Hy-Analgesia treatment (r = .34,
P < .05). However, in order to evaluate how changes
in experienced pain and distress scores, as induced by
Fz
-20
-10
0
10
Cz
-20
-10
0
10
Pz
-20
-10
0
10
-99 0 99 198 297 396 495
ms
V
Fz
-20
-10
0
10
Cz
-20
-10
0
10
Pz
-20
-10
0
10
-99 0 99 198 297 396 495
Fz
-20
-10
0
10
Cz
-20
-10
0
10
Pz
-20
-10
0
10
-99 0 99 198 297 396 495
W-Pain W-Analgesia
Hy-Pain
Hy-Analgesia P.Hy-Analgesia
High Hypnotizables Medium Hypnotizables Low Hypnotizables
Fig. 3. Somatosensory event-related potentials (ERPs) to painful standard stimuli averaged across frontal, central, and parietal midline sites in high,
medium and low hypnotizable subjects during Waking, Hypnosis and Post-Hypnosis conditions for Pain and Analgestia treatments.
204 V. De Pascalis et al. / Pain 134 (2008) 197­208
analgesic treatments, were related with changes in N140
and P200 amplitudes, Pearson's correlation coefficients
were calculated. Change scores were calculated by subtracting
scores obtained during W-Analgesia from those
obtained during W-pain, and scores obtained during
Hy-Analgesia and P.Hy-Analgesia from those during
Hy-Pain. The reduction of experienced pain was significantly
correlated with the reduction of midline frontal
N140 amplitude during Hy-Analgesia and P.Hy-Analgesia
treatments (r = .40, P < .05 and r = .43, P < .01,
respectively). Similarly, the reduction of experienced distress
was significantly correlated with the reduction of
midline frontal N140 amplitude during Hy-Analgesia
and P.Hy-Analgesia treatments (r = .43, P < .01 and
r = .44, P < .01, respectively). No significant correlations
were found between pain or distress reductions
and P200 amplitudes of the ERPs.
4. Discussion
Self-report data displayed differential patterns of
responding with MH individuals showing a moderately
lower level of pain during hypnosis as compared to
waking condition, and LH individuals showing little
differentiation between the waking, hypnosis, and posthypnosis
conditions. In contrast, HH individuals displayed
significantly lower pain and distress scores under
Hy-Analgesia and P.Hy-Analgesia as compared to WPain
and Hy-Pain conditions. These results are consistent
with a number of studies showing that HH individuals
are more responsive to hypnotic analgesia
suggestions [2,10,14,47,48,50,51,61]. Moreover, the finding
that HH subjects experienced significant pain and
distress reductions during P.Hy-Analgesia as compared
to Hy-Analgesia condition supports the assumption that
the magnitude of responses to suggestions is enhanced
during a more intense individuaľs hypnotic involvement
[28,53]. The findings that pain and distress ratings did
not differ between W-Pain and Hy-Pain treatments
among hypnotizability groups (see Figs. 1 and 2) suggest
that differential respondings to painful stimuli were due
to the influence of the hypnotic suggestions rather than
to an effect of the hypnotic induction per se. In fact, in
the present study, MH and LH individuals were less
responsive to the hypnotic suggestions than the HH
individuals, although the former during Hy-Analgesia
and P.Hy-Analgesia reported smaller pain and distress
ratings than during W-Pain and W-Analgesia treatments
(see white columns on Figs. 1 and 2).
It is important to note that, in the present study, in
line with the recommendation of [38], the impact of hypnosis
on the response to suggestion was tested by delivering
the exact same suggestion to subjects in and out of
hypnosis. The findings indicate that in a condition of
deep hypnosis (as it is assumed to be the case of posthypnotic
condition), the analgesic effect of suggestions
was amplified in comparison with both the waking
non-hypnotic condition and the standard hypnotic
condition.
In terms of retrospective pain and distress ratings, the
present results indicate that HH subjects, five minutes
after the conclusion of hypnosis, over-evaluated the
effect of pain reduction they experienced during HyAnalgesia
and P.Hy-Analgesia treatments. In contrast,
for W-Pain, W-Analgesia, and Hy-Pain conditions,
these subjects remembered similar pain and distress levels
to those they concurrently experienced with
stimulation.
It is known that negative affect increases acute pain
report and has been implicated with the level of maximum
pain that is recalled and in the extent to which
the memory of pain becomes distorted across time
[24,25,56]. The more pronounced pain and distress ratings,
as recollected for waking and hypnosis-pain conditions,
provide an experiential context through which the
painful stimulus is processed and later better recalled.
This interpretation is consistent with research suggesting
that emotionally arousing or stressful experiences are
generally well remembered, and that pain processing
and recall are highly dependent upon contextual cues
associated with the painful experience [45]. It may be
that the more pronounced memory distortions of pain
and distress, as observed during hypnotic analgesic
treatments, can be explained as a proactive effect of
the reduced emotional distress during these hypnosis
treatments.
In terms of SERP's changes, the most striking feature
is a drop in amplitude for both the N140 and P200 peak
amplitudes, observed for the HH individuals, over frontal
and central recordings during both Hy-Analgesia
and P.Hy-Analgesia treatments. The expected more pronounced
peak amplitude reduction during P.Hy-Analgesia,
as compared to Hy-Analgesia treatment, was not
observed. This apparent lacking difference may be due
to a floor effect since the drop in amplitude for N140
and P200 components, during analgesic treatments in
hypnosis, was quite large (about 5 lV, see Fig. 3). For
the MH and LH subjects, no significant amplitude
changes were observed during analgesic treatments in
hypnosis as compared to painful treatments.
There is experimental evidence that the somatosensory
N140 peak increases with selective attention
[18,19,23,41,43]. However, the drop in amplitude, that
we observed for the N140 component during Hy-Analgesia
and P.Hy-Analgesia treatments in HH subjects,
cannot be univocally seen as a reflection of reduced
arousal. This is because no effect of a hypnotic induction
alone, usually associated with a reduced level of arousal,
was observed on N140 amplitude (i.e., the N140 amplitude
during Hy-Pain was not smaller than during WPain).
Considering that during Hy-Analgesia and
P.Hy-Analgesia pain and distress reductions were found
V. De Pascalis et al. / Pain 134 (2008) 197­208 205
significantly correlated with the reductions of frontal
N140 amplitude, it can be suggested that, during hypnotic
analgesic treatments, the drop in N140 amplitude
is the product of the modulation of pain experience
associated with the obstructive nature of the ``glove
analgesia'' suggestion. This suggestion implies a modification
or reinterpretation of the sensation as non-painful
rather than a distraction from the pain sensation
since the oddball paradigm requires the subject to attend
the stimulus.
A variety of studies suggest that the later positive
components of the SERPs are related to the painfulness
of the stimuli and that these components may even be
lacking when the stimuli are not experienced as painful
[4,26]. Thus, assuming that the P200 component is a
pain specific component of the SERPs [4], the reduced
P200 peak, observed in the present study during hypnotic
and posthypnotic analgesia treatments, can be
seen as reflecting the operation of an inhibitory mechanism
in the pain processing system. This inhibitory process
may be more of a global than specific nature since it
involves attentional and pain-specific components of the
SERPs. A similar inhibitory effect, involving negative
and positive ERP components, has been also reported
in previous hypnosis studies using obstructive imagery
of incoming visual [11,54,55] and somatosensory stimuli
[14,15,48].
In the present study, both N140 and P200 components
displayed a more fronto-central focus with a
greater modulation of their peak amplitudes as a function
of hypnotic involvement. This observation parallels
the findings reported by [48] who reported two negative
components of the ERPs maximal at vertex (N140 and
P250) in response to electric stimulation and suggested
to reflect the activity of cingulate and subcortical areas.
Thus, the present findings appear in agreement with the
hypothesis that hypnotic suggestions, in HH individuals,
influence the somatosensory areas of the cortex. The
statement that primary somatosensory cortex plays a
leading role in pain perception and is important for
the discrimination and modulation of various aspects
of pain is derived by animal and human studies (see
[6,20]). This modulation consists in the inhibition of
the sensory areas through cortico-thalamic descending
projections.
In conclusion, our findings that obstructive suggestions
reduced both the early N140 and the later P200
peaks in high susceptible subjects under hypnosis support
the hypothesis that hypnosis procedure can affect
earlier and later stages of stimulus processing (see
Fig. 3). The early effect (N140) may be related to differences
in arousal in the relevant circuitry that may be
mediated by structures such as the parts of the ascending
colinergic reticular arousal system [49] or the intralaminar
nuclei of the thalamus [35]. The later effects (i.e., the
reduced P200 peak amplitude for the hypnotic analgesic
conditions) can be seen as a reflection of the reduced
activity in the anterior cingulate or prefrontal cortex,
known to be activated as parts of the aversive response
[7,9,17].
Although this study provides experimental evidence
that hypnosis is a valid tool for pain reduction in
HH subjects, the interpretation of the present finding
is limited by the potential emotional effect of the oddball
stimuli involving painful shocks. The parameters
used in this study for the oddball target stimuli suggest
that they were likely to be perceived as quite painful
(i.e. a pair of standard stimuli which were adjusted to
be close to, or at, tolerance level). This implies that
the response to standard stimuli may actually involve
an anticipatory component that may induce some anxiety
associated with the risk of receiving a more painful
shock. The analgesic effect reported may, therefore,
reflect both a direct effect on the response to standard
stimuli as well as an indirect effect due to the reduction
in the anxiety associated to the very painful target
stimuli. Therefore, to avoid this potential emotional
effect, future oddball ERP studies on pain modulation
in hypnosis should use standard and target stimuli with
similar intensity and randomized interstimulus
intervals.
References
[1] Babiloni C, Babiloni F, Carducci F, Cincotti F, Rosciarelli F,
Arendt-Nielsen L, et al. Human brain oscillatory activity phaselocked
to painful electrical stimulations: a multi-channel EEG
study. Hum Brain Mapp 2002;15:112­23.
[2] Barabasz A, Watkins JG. Hypnoterapeutic techniques 2E. New
York: Brunner-Loutledge; 2005. p. 192­194.
[3] Becker DE, Yingling CD, Fein G. Identification of pain, intensity
and P300 components in the pain evoked potential. Electroencephalogr
Clin Neurophysiol 1993;88:290­301.
[4] Becker D, Haley DW, Urena VM, Yingling CD. Pain measurement
with evoked potentials: combination of subjective ratings,
randomized intensities, and long interstimulus intervals produces
a P300-like confound. Pain 2000;84:37­47.
[5] Blom JL, Anneveldt M. An electrode cap tested. Electroencephalogr
Clin Neurophysiol 1982;54:591­4.
[6] Bushnell MC, Duncan GH, Hofbauer RK, Ha B, Chen J, Carrier
B. Pain perception: is there a role for primary somatosensory
cortex? Proc Natl Acad Sci USA 1999;96:7705­9.
[7] Carter CS, Botvinick MM, Cohen JD. The contribution of the
anterior cingulate cortex to executive processes in cognition. Rev
Neurosci 1999;10:49­57.
[8] Chaves J, Dworkin S. Hypnotic control of pain: historical
perspectives and future prospects. Int J Clin Exp Hypn
1997;45:356­76.
[9] Corbetta M. Frontoparietal cortical networks for directing
attention and the eye to visual locations: identical, independent,
or overlapping neural systems? Proc Natl Acad Sci USA
1998;95:831­8.
[10] Danziger N, Fournier E, Bouhassira D, Michaud D, De Broucker
T, Santarcangelo E, et al. Different strategies of modulation can
be operative during hypnotic analgesia: a neurophysiological
study. Pain 1998;75:85­92.
206 V. De Pascalis et al. / Pain 134 (2008) 197­208
[11] De Pascalis V. Event-related potentials during hypnotic hallucination.
Int J Clin Exp Hypn 1994;XLII:39­55.
[12] De Pascalis V, Cacace I, Massicolle F. Perception and modulation
of pain in waking and hypnosis: functional significance of phaseorder
gamma oscillations. Pain 2004;112:27­36.
[13] De Pascalis V, Chiaradia C, Carotenuto E. The contribution of
suggestibility and expectation to placebo analgesia phenomenon
in an experimental setting. Pain 2002;96:393­402.
[14] De Pascalis V, Magurano MR, Bellusci A. Pain perception,
somatosensory event-related potentials and skin conductance
responses to painful stimuli in high, mid, and low hypnotizable
subjects: effects of differential pain reduction strategies. Pain
1999;83:499­508.
[15] De Pascalis V, Magurano MR, Bellusci A, Chen ACN. Somatosensory
event-related potentials and autonomic activity to varing
pain reduction cognitive strategies in hypnosis. Clin Neurophysiol
2001;112:1475­85.
[16] De Pascalis V, Bellusci A, Russo PM. Italian norms for the
Stanford hypnotic susceptibility scale, form C. Int J Clin Exp
Hypn 2000;48:315­23.
[17] Desimone R, Duncan J. Neural mechanisms of selective visual
attention. Annu Rev Neurosci 1995;18:193­222.
[18] Desmedt JE, Tomberg C. Mapping early somatosensory evoked
potentials in selective attention: critical evaluation of control
conditions used for titrating by difference the cognitive P30, P100
and N140. Electroencephalogr Clin Neurophysiol 1989;74:321­46.
[19] Eimer M, van Velzen J. Crossmodal links in spatial attention are
mediated by supramodal control processes: evidence from eventrelated
potentials. Psychophysiology 2002;39:437­9.
[20] Ergenzinger ER, Glasier MM, Hahm JO, Pons TP. Cortically
induced thalamic plasticity in the primate somatosensory system.
Nat Neurosci 1998;1:226­9.
[21] Farvolden P, Woody EZ. Hypnosis, memory, and frontal executive
functioning. Int J Clin Exp Hypn 2004;52:3­26.
[22] Feine JS, Lavigne GJ, Dao TTT, Morin C, Lund J. Memories of
chronic pain and perceptions of relief. Pain 1998;77:137­41.
[23] Garcia-Larrea L, Lukaszewicz A-C, Mauguiere F. Somatosensory
responses during selective spatial attention: the N120­N140
transition. Psychophysiology 1995;32:526­37.
[24] Gedney JJ, Logan H. Memory for stress-associated acute pain. J
Pain 2004;5:83­91.
[25] Gedney JJ, Logan H, Baron RS. Predictors of short-term and
long-term memory of sensory and affective dimensions of pain. J
Pain 2003;4:47­55.
[26] Gibson SJ, LeVasseur SA, Helme RD. Cerebral event-related
responses induced by CO2 laser stimulation in subjects suffering
from cervico-brachial syndrome. Pain 1991;47:173­82.
[27] Hilgard ER. Divided consciousness: multiple controls in human
thought and action. New York: Wiley; 1977. p. 62­86.
[28] Hilgard ER. Suggestibility and suggestions as related to hypnosis.
In: Schumaker J, editor. Human suggestibility. New
York: Routledge; 1991. p. 37­58.
[29] Jamison RN, Sbrocco T, Parris WCV. The infuence of physical
and psychological factors and accuracy for memory for pain in
chronic pain patients. Pain 1989;37:289­94.
[30] Jensen M, David R, Patterson DR. Hypnotic treatment of chronic
pain. J Behav Med 2006;29:95­124.
[31] Jensen MP, Karoly P, Braver S. The measurement of clinical pain
intensity: a comparison of six methods. Pain 1986;27:117­26.
[32] Kihlstrom JF. Hypnosis: a sesquicentennial essay. Int J Clin Exp
Hypn 1992;40:301­14.
[33] Kihlstrom JF. Convergence in understanding hypnosis? Perhaps,
but perhaps not quite so fast. Int J Clin Exp Hypn
1997;45:324­32.
[34] Kihlstrom JF, Schacter DL. Functional disorders of autobiographical
memory. In: Baddeley AD, Wilson BA, Watts FN,
editors. Handbook of memory disorders. Chichester, UK: Wiley;
1995. p. 337­64.
[35] Kinomura S, Larsson J, Gulyas B, Roland PE. Activation by
attention of the human reticular formation and thalamic intralaminar
nuclei. Science 1996;271:512­5.
[36] Kirk RE. Experimental design: procedures for the behavioral
sciences. Belmont: Brooks; 1968.
[37] Kirsch I. Suggestibility or hypnosis: what do our scales really
measure? Int J Clin Exp Hypn 1997;45:212­25.
[38] Kirsch I, Mazzoni G, Montgomery GH. Remembrance of
hypnosis past. Am J Clin Hypn 2007;49:171­8.
[39] Linton SJ, Melin L. The accuracy of remembering chronic pain.
Pain 1982;13:281­5.
[40] Mathias BJ, Dillingham TR, Zeiger DN, Chang AS, Belandres
PV. Topical capsaicin for neck pain: a pilot study. Am J Phys Med
Rehabil 1995;74:39­44.
[41] Michie PT. Selective attention effects on somatosensory eventrelated
potentials. Ann NY Acad Sci 1984;425:250­5.
[42] Morgan AH, Hilgard JR. The Stanford hypnotic clinical scale for
adults. Am J Clin Hypn 1978­1979;21:134­47.
[43] Na¨a¨ta¨nen R, Michie PT. Early selective-attention effects on the
evoked potential: a critical review and reinterpretation. Biol
Psychol 1979;8:81­136.
[44] Orne M. Hypnosis, motivation, and compliance. Am J Psychol
1966;122:721­6.
[45] Pearce SA, Isherwood S, Hrouda D, Richardson PH, Erskine A,
Skinner J. Memory and pain: tests of mood congruity and state
dependent learning in experimentally induced and clinical pain.
Pain 1990;43:187­93.
[46] Posner MI, Petersen SE. The attention system of the human brain.
Annu Rev Neurosci 1990;13:25­42.
[47] Rainville P, Hofbauer RK, Paus T, Duncan GH, Bushnell MC,
Price DD. Cerebral mechanisms of hypnotic induction and
suggestion. J Cogn Neurosci 1999;11:110­25.
[48] Ray WJ, Keil A, Mikuteit A, Bongartz W, Elbert T. High
resolution EEG indicators of pain responses in relation to
hypnotic susceptibility and suggestion. Biol Psychol
2002;60:17­36.
[49] Robbins TW. Arousal systems and attentional processes. Biol
Psychol 1997;45:57­71.
[50] Sharav Y, Tal M. Focused analgesia and generalized relaxation
produce differential hypnotic analgesia in response to ascending
stimulus intensity. Int J Psychophysiol 2004;52:187­96.
[51] Sharav Y, Tal M. Focused hypnotic analgesia: local and remote
effects. Pain 2006;124:280­6.
[52] Spanos NP. Hypnotic behavior: a social psychological interpretation
of amnesia, analgesia and ``trance logic''. Behav Brain Sci
1986;9:449­67.
[53] Spiegel D. A definition without a definition. Cont Hypn
1994;11:151­2.
[54] Spiegel D, Barabasz AF. Effects of hypnotic instructions on P300
event-related potential amplitudes: research and clinical implications.
Am J Clin Hypn 1988;31:11­7.
[55] Spiegel D, Cutcomb S, Ren C, Pribram K. Hypnotic hallucination
alters evoked potentials. J Abnorm Psychol 1985;94:249­55.
[56] Tasmuth T, Estlanderb AM, Kalso E. Effect of present pain and
mood on the memory of past postoperative pain in women treated
surgically for breast cancer. Pain 1996;68:343­7.
[57] Traina F. Scala per il Profilo Stanford di Suscettibilita` Ipnotica:
Forma C di A. M. Weitzenhoffer e E. R. Hilgard (adattamento
Italiano) [Stanford Hypnotic Susceptibility Scale: Form C, by A.
M.Weitzenhoffer and E. R. Hilgard (Italian adaptation)]. Firenze,
Italy: Organizzazioni Speciali; 1976.
[58] Vasey MW, Thayer JF. The continuing problem of false positives
in repeated measures ANOVA in psychophysiology: a multivariate
solution. Psychophysiology 1987;24:479­86.
V. De Pascalis et al. / Pain 134 (2008) 197­208 207
[59] Weitzenhoffer AM, Hilgard ER. Stanford hypnotic susceptibility
scale: form C. Palo Alto, CA: Consulting Psychologist Press; 1962.
[60] Woody EZ, Bowers KS. A frontal assault on dissociated control. In:
Lynn SJ, Rhue JW, editors. Dissociation: clinical theoretical and
research perspectives. New York: Guildford Press; 1994. p. 52­79.
[61] Zachariae R, Andersen OK, Bjerring P, Jorgensen MM,
Arendt-Nielsen L. Effects of an opioid antagonist on pain
intensity and withdrawal reflexes during induction of hypnotic
analgesia in high- and low-hypnotizable volunteers. Eur J Pain
1998;2:25­34.
208 V. De Pascalis et al. / Pain 134 (2008) 197­208